© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3228
    Prev
    Next
    INDEX
    Explanations

    UPS

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Perkins
    -0.10
    è¢į
    -0.10
    OMIC
    -0.09
    CAPE
    -0.09
    íĪ¬
    -0.09
    æijĦ
    -0.09
    iton
    -0.09
    fully
    -0.09
    æĤ«
    -0.09
    fulness
    -0.09
    POSITIVE LOGITS
    ilon
    0.10
    chrift
    0.09
    olini
    0.09
    rück
    0.09
    åľ¨æĹ¥æľ¬
    0.09
    ILON
    0.09
    mere
    0.08
    èİ«æĸ¯
    0.08
    çŀİ
    0.08
    arÃŃa
    0.08
    Activations Density 0.010%

    No Known Activations